Category: Tech

EmptyWheel on Artificial Intelligence

Worth your time

I highlighted the excellent 4-part series What We Talk About When We Talk About AI by EmptyWheel in the most recent Sunday Morning Reading column. I’m linking to it again, because it is not only excellent in scope and content, but also depth and context that will be worth referring to as we continue to plunge, or be plunged, into a world increasingly confounded and defined by AI.

Peter herrmann 9_fk2tp kla unsplash.

As I said, it’s a four part series. Here are links to each piece:

A NORMAL PERSON’S EXPLAINER ON WHAT GENERATIVE AI IS AND DOES

THE OTHER HALF OF THE AI RELATIONSHIP

PROTEINS, FACTORIES, AND WICKED SOLUTIONS

LLMS ARE LEAD

Repeating for my own emphasis something I’m saying lately, the pendulum of opinion about AI is swinging widely of late. The more ubiquitous it becomes the more we seeing opinions shift. Sometimes away from using it, but also seeing some who have been negative about using AI, reevaluating it as well.

That all seems natural to me given where we are with this new technology. Essentially in its infancy. A well funded infancy perhaps, but still teething, and shitting the bed on occasion.

Bookmark this series or save it where you will. I have a feeling it will be essential context in the short and long terms of this era we’re in.

(Image from Peter Herrman on Unsplash.)

You can also find more of my writings on a variety of topics on Medium at this link, including in the publications Ellemeno and Rome. I can also be found on social media under my name as above. 

Things Apple Won’t Fix In The Next Round of OS 26 Updates

I’d bet money these issues persist after new OS releases

Apple is announcing new iPhones, Apple Watches and possibly other things today. They’ll also announce the release date for the operating systems for all of those new devices and previous generations already in use. All of the focus is going to be on everything new, improved, changed and Liquid Glassed.

55b31eb4ba867b00263c62f538280311 XL.

I’m sure I’ll write about some of that down the road. For today I’m writing about the things I’m thinking Apple won’t address in any of the new operating systems. These are lingering issues that in some cases have not been corrected over the life of multiple operating systems. As much as I get excited about a new software release I remain frustrated year after year when Apple steps forward while dragging its feet on correcting existing problems.

Believe me, I get that Apple has created such a large ecosystem that it is impossible to pay attention to every detail and correct every mistake. I get it, but I don’t sympathize with it. Introducing new operating systems pumped out on what seems like an impossible to meet annual cycle creates other issues that need to be addressed and it is apparently too easy to just let the older ones go unattended. And from what I’ve seen from developers I follow, there are going to be a host of new issues that are going to need to be resolved over the next year. Or longer. The longer you let an existing issue hang around without attention, the more the “control the whole widget” promises seem not quite in control.

Certainly it’s a question of priorities. How the priorities are set makes me question the priorities of those that set them.

Some of these issues I’ve highlighted before, others I have not. Here are a few that continue my questioning of Apple’s priorities. I’m sure there are some that others experience that fall into the same category of hoping Apple pays attention to them someday.

Syncing

I’ve seen a decrease in syncing issues with iCloud in general. That’s a good thing. But Apple still insists on syncing Photos on its own schedule. Whether it’s battery life optimization or system optimization the delay is never optimized for the user experience.

Shared Reminder Notifications

I’ve written about this before. The fact that this one hasn’t been corrected since Shared Reminders became a thing undercuts all of the very good work Apple has done with the Reminders app in general.

Bluetooth Weirdness

I don’t know how else to describe this other than it’s just weird. There are times when Bluetooth connectivity works as designed with my AirPods and other peripherals. There are times when things just go bonkers especially with AirPods. Things will work as designed for days on end, then they won’t. AirPods switch between devices as designed, but then stop doing so. Adaptive Listening will work well sometimes and sometimes not. Connectivity with my car works well and then it doesn’t. My car isn’t smart by any stretch of the imagination and gets no software updates so it makes sense to me that it’s a Bluetooth issue on the iPhone. Note that it happens at random times unrelated to iOS updates or those mysterious AirPod firmware updates. The Magic Trackpad works 98% of the time. Then not. It’s weird.

Shortcuts

It’s become either a legend or a joke that every time Apple releases a new update (full release or point release) something goes wrong with Shortcuts. Either way it’s a problem for “the future of automation.” I particularly love it when a Shortcut just hangs with no indication that there’s a problem after working well for days on end. This inconsistent problem is consistent.

The App Store

Apple’s biggest retail outlet is the slowest loading and buggiest piece of software it has ever designed. It’s also designed abysmally.

Focus Modes

I use very few Focus Modes. I do not share them between devices. Sometimes they work. Sometimes they don’t. I’ve never set a Sleep Focus on my iMac. Yet one mysteriously appears at random during the day about twice a month. The Focus I’ve set for listening to Podcasts inconsistently fails to kick in.

Mail Rules

I think Apple has just given up on Mail Rules because the ones I’ve built just don’t work anymore.

Apple News

I enjoy that on my Mac I can choose to open an article in Safari or that I can open a webpage in Apple News. I don’t understand why that can’t happen on the iPhone.

Notification Summaries

I long ago turned Notification Summaries off because they were just confusing and caused more work than just dealing with the Notification. I recently turned them back on to check and see if there had been any improvement. There has not.

As I said, I don’t anticipate Apple fixing any of the above with all of the new ’26 releases. I’ve learned to work around them the same way I work around the passenger side backdoor on my car won’t open from the inside and I have to open it from the outside whenever I’ve got a back seat passenger. It’s a walk around workaround. It’s not ideal. And each time I have to do it I’m reminded just how imperfect things are and can be. I’m sure it will diminish the value of the car when I do decide to trade it in, but it’s already reduced the value of it in my mind.

I feel the same way when Apple doesn’t address longstanding issues that I and others have brought to their attention. Here’s hoping Apple addresses some of these issues. Here’s hunching they won’t.

You can also find more of my writings on a variety of topics on Medium at this link, including in the publications Ellemeno and Rome. I can also be found on social media under my name as above. 

Instagram Shows Up Very Late to the iPad Party

What’s the point and who cares?

The folks at Meta must have something up their sleeve. The reason I say that is they have finally, after all of these years released an iPad version of the app, long after most folks just figured it would never happen. Other than speculating on what might be behind the late to the party move, at this point it begs the question, Who cares?

Reels with Comments 16_9.

I’m sure plenty do care. I’m not one of them. Instagram is one of a few apps that I begrudgingly use. Begrudgingly because I hate it. I use it because my family on all sides continues to do so and it’s a way to keep up with grandkids, nieces, nephews, and other family news. But I honestly despise that I have to. Believe me I’ve tried to wean them off onto other apps and services, but it never sticks.

Every time I do open Instagram I have to block somewhere between 5 and 10 spam accounts (too often porn or ridiculous come ons.) And of course the algorithm doesn’t show me what I want to see, but what it wants me to see. There’s even an increased sense of desperation from both Instagram and Facebook sending out notifications telling me someone replied, is waiting for my reply, or commented on something I haven’t seen yet. It’s like begging in the street. Apologies to those who might actually need to do so.

Sure I could turn off the notifications, but sadly, that’s the least worst way to use the app to keep up with family happenings.

I’d say that since it took 15 years for Meta to finally roll this out that perhaps the adolescents in charge finally are growing up. But then, there are those porn accounts that pop up with the frequency like prepubescent zits.

I won’t be putting it on my iPad. It’s troubling enough that I still have it on my iPhone. And as I watch the over excited coverage rolling in, I haven’t seen any image of the iPad version that makes it look the least bit appealing. It’s like Meta didn’t really care based on what I’ve seen so far.

So, Instagram is on the iPad. What’s the point and who cares?

You can also find more of my writings on a variety of topics on Medium at this link, including in the publications Ellemeno and Rome. I can also be found on social media under my name as above. 

Blood Glucose Monitoring on the Apple Watch Is Probably A Long Way Off

The annual hope and hype cycle

Every year as we near Apple’s iPhone and Apple Watch announcements rumors circulate and recirculate about Apple including a blood glucose monitor for the Apple Watch. It would be great if that happens someday. In my view it’s probably a lot further off than most of those hoping it happens want it to be. Here’s why I think that.

I’m a Type 2 diabetic. Checking my blood sugars has been a part of my life for 14 years or so. I’m painfully familiar with the prick your finger method and relied on it until a new endocrinologist prescribed a Dexcom G6 sensor. I’m now using a Dexcom G7 sensor. I don’t use an automatic insulin pump, just the sensors for monitoring via my iPhone and Apple Watch. I treat my condition with injected medication, both insulin and GLP-1.

The Dexcom sensors (both generations) were indeed a plus in my life as they provided a better way of monitoring, allowing me to better gauge diet and exercise, which will always be the biggest part of any diabetic’s regime.

But neither Dexcom model was a fire and forget it solution. There are times when I have to check my blood sugar with a finger prick because there are so many variables that can affect your glucose levels. The Dexcom system does provide a method to calibrate and recalibrate based on actual readings from a finger prick. I experience periods when I don’t need to recalibrate often and periods when I do.

If I’m traveling for a more than a day or so I have to pack my kit for finger pricks, because travel can be one of those variables. An inadvertent bump of the sensor on my arm can throw readings off for the remaining life of that sensor. (You have to change to a new sensor every 10 days.) AND to be frank, the tech is still far from perfect. Sensors fail occasionally and, as is the case with all tech that fails, it happens at the most inappropriate time.

Dexcom’s sensors are classified as a minimally invasive medical device. The device inserts a small filament under the skin, and yes it feels like a small needle insertion. That filament takes its reading from the interstitial fluid between cells. It does not read directly from the bloodstream. The device is held on by an adhesive patch.

When Dexcom introduced its Apple Watch app it was a decided monitoring improvement over the iPhone App which I viewed via a widget on my iPhone Home Screen. The iPhone widget far too frequently needed a touch of the screen which opened the app to update the reading.

The Apple Watch app offers a complication that displays data more consistently, needing less frequent touches to refresh the readings. (The complication is visible in the photo above showing a reading of 157.) That said, the Apple Watch tends to disconnect too easily from the sensor, defaulting back to displaying readings from the iPhone when that occurs.

I also find that the Apple Watch app reading is more current than what the iPhone is showing, but you do still need the iPhone app to communicate readings with your endocrinologist.

In some instances I’ll need to reboot both devices to reconnect things up on both my iPhone and Apple Watch. Dexcom technicians have told me these disconnections have to do with Bluetooth connection issues and are subject to the same vagaries and variables that exist with many other Bluetooth connections. I also need to be aware of which arm I have the sensor attached. I wear my Apple Watch on my left arm and if the sensor is on my right arm there will be more frequent disconnections. I don’t sleep with my Apple Watch on, so waking in the morning requires a waiting period for the watch app to reconnect.

The goal behind the push to include blood glucose monitoring on an Apple Watch (or any other smart device) is obviously to cut down on the need for finger pricks and possibly the use of minimal invasive medical devices. But also, in theory an on device sensor communicating directly with watchOS should show readings more consistently and be much less subject to the vagaries of Bluetooth connectivity.

I certainly can’t prejudge any new system or technology that hasn’t surfaced yet. Certainly I’m one who hopes Apple or some other company can eventually tackle this issue and provide a workable solution. That said, convenient as this annual rumor always seems to be, I believe blood glucose monitoring on the Apple Watch is a lot further off than most hope or think it might be. I’m also reasonably sure that finger pricks will still be needed if for no other reason than to calibrate sensor readings with actual blood glucose levels.

You can also find more of my writings on a variety of topics on Medium at this link, including in the publications Ellemeno and Rome. I can also be found on social media under my name as above. 

AI Smoothing The Rise of Fascism

The proliferation of greater fools

This a quick link to a great piece by David Todd McCarty called The AI Ponzi Scheme And The Greater Fool. Go read it.

Annie spratt JcIPxdGuE_Y unsplash.

Although the piece is richer and deeper than the point I’m singling out, McCarty ties the proliferation of Artificial Intelligence to rise of fascism, neatly summed up by this quote from the piece:

“Once we get to a point where nothing is trustworthy, an authoritarian regime is free to lie with impunity.”

I concur.

I will also go a bit further and say we’ve already reached that point. It’s how we handle it now that we’re here that matters.

Again, go read the piece.

Image from Annie Spratt on Unsplash.

Sunday Morning Reading

The ant hill of humanity

Crazy travel rhythms this summer. Spending time at the lake this weekend. The good thing about lake time is there’s time to do some reading. Here’s some good stuff I stumbled onto, worth sharing for this week’s edition of Sunday Morning Reading.  Quite a bit revolving around Artificial Intelligence and other mind games. There’s also ants.

For some inexplicable reason defining what it means to be an American has actually become a chore these days. It shouldn’t be. Kieran Healy has written a piece simply titled American that recounts his thoughts and feelings on becoming an American citizen. Well worth your time, espeically in these crazy times.

“Memory isn’t linear; it’s relational.” That’s the thought NatashaMH leaves us with in her piece The Mind’s Mischief. The mind is indeed a curious thing.

Matteo Wong says the AI Doomers Are Getting Doomier. I don’t know about you, but if we’re all doomed at the hands of AI (does AI have hands?) human intelligence never really advanced as far as I thought it did. Or maybe we just hit the ceiling.

Speaking of AI doom, Charlie Warzel wonders why one of the impacts of AI it to make us feel like we’re losing it in  AI Is A Mass-Delusion Event. I get the points and they’re well made. Referring back to my comment from the previous entry, if we’re such easy marks for this kind of delusion… well…we are such easy marks.

David Todd McCarty argues why we should resist AI with ecclesiastical fervor, especially those who create for a living. Check out The Moral Failure Of Using AI In Your Art.

Reece Rogers is marking yet another change brought about by AI. Take a look at The AI-Powered PDF Marks The End of An Era.

Barry Betchesky tells us that It Took Many Years And Billions of Dollars, But Microsoft Finally Invented A Calculator That Is Wrong Sometimes. You read that right. The money quote is:

“But now we have Microsoft apparently determining that ‘unpredictability’ was something that some number of its customers wanted in their calculators.”

Rounding out this collection of links on AI, is another article by NatashaMH where she says instead of Fearing the Machinery, Interrogate The Mindset. Excellent piece. The underlying current is something I’ve been thinking about a lot. We’re creating these machines in our own images. Or at least the images we imagine of ourselves. Humans are far too human, even when we look past or try to accelerate beyond our humanity.

One of the joys of spending time in the great outdoors is that it reminds you we’re not the only intelligent species on the planet. Although as the theme of this week’s reading has emerged, we might want to reevaluate that, just not with Microsoft’s math tools. On another front, in politics it’s certainly easy to argue for a reevaluation. Kate Knibbs takes a swipe at it in a look at how Government Staff Cuts Have Fueled An Ant-Smuggling Boom.

I told you there’d be ants.

If you’re interested in just what the heck Sunday Morning Reading is all about you can read more about the origins of Sunday Morning Reading here. If you’d like more click on the Sunday Morning Reading link in the category column to check out what’s been shared on Sunday’s past. You can also find more of my writings on Medium at this link, including in the publications Ellemeno and Rome.

Who Controls History If AI Is In The Mix

Time in a bottle of bits

One of the scariest things about this insane period we’re living through is the attempt by those in power to rewrite, alter, or just get rid of history they don’t like. Whether it’s banning books, changing curriculums, forcing the closures of libraries, or what museums can display, I find it a cowardly, yet effective way to hide heads in the sand, bury the sins of the past, and admit we’re actually ashamed of ourselves.

Peter herrmann 9_FK2Tp kLA unsplash.

I know this because I have lived this. My early education certainly tilted the American narrative towards the mythology of the Old South. It wasn’t until I left home, and got involved in the theatre that I discovered just how much I hadn’t learned, how much more I needed to, and how the future depends on the past, no matter how complicated it was.

Fortunately the information was there. It was up to me to do the work.

What happens when it’s not there? Or it’s wrong?

I find It hard to imagine that large chunks of the world’s history can be erased, growing up in an era when my access to it seemed to expand exponentially. But it’s been tried before. It’s succeeded with entire generations of populations. Now we’re facing the very real possibility of it happening again in this digital age with the aid of Artifical Intelligence.

There’s an interesting piece by Benji Edwards in Ars Technica about a college student who trained a small AI model that he called TimeCapsuleLLM on Victorian texts. During his experiments his time capsule spit out some actual history he didn’t know about real protests during the era. He checked into the info and the LLM was indeed accurate.

At first glance, that feels like a very positive AI story. Discovering lost history is a good thing. However, with the way I understand AI training it all depends on what data it’s trained on. That leaves things up to who controls the training data. Leave out, change or bias the historical record and…

Well, you can see the problem.

Elon Musk has already hinted at this kind of manipulation. I’m sure there are others thinking the same. They say history repeats itself. Actually history doesn’t. Humans do. History is just the record of the repetition. Humans just use newer and different tools to mold the past into something more comfortable. I may be mistaken, but I think history, in the long run, also proves that never really works out.

Correcting and rewriting history is not for the faint of heart. But when there is no heart, there’s a problem.

Time machines and time travel have always been fraught with danger in the history of science fiction. So has Artificial Intelligence. I’m reasonably sure we’re not smart enough to walk whatever fine lines might exist in a future when the past can be more easily manipulated. We haven’t been in the past when the erasing was harder. But I am dead certain we’re going to be facing this unreal reality.

Again.

Just with newer methods.

Without anything resembling Artficial Intelligence, we’ve managed to forget, alter, or set aside many of the horrible lessons of human history. Why should any new tool we create be any different? I’m sure these AI geeks think they can strip ego and emotion out of these robots they are building.

I doubt they will ever remove hubris.

(Image from Peter Herrman on Unsplash.)

You can also find more of my writings on a variety of topics on Medium at this link, including in the publications Ellemeno and Rome. I can also be found on social media under my name as above. 

Craig Hockenberry’s Take on the Why of Liquid Glass

Is Liquid Glass an edge case?

There is a quote largely attributed to Robert F. Kennedy. No not the one rampaging through the U.S. healthcare system. It goes “some men see things as they are and ask why. I dream things that never were, and ask why not?” But, it actually came first from George Bernard Shaw in his play Back to Methuselah. 

Apple WWDC25 Liquid Glass hero 250609_big.jpg.large_2x.

The text gets mangled often enough as does the attribution. Regardless, the point is made. In my experience as a theatre producer/director/designer/playwright the biggest part of the game is the “dreaming” part. The next hurdle is finding a way to turn that into reality. The dreams often come when least expected, occasionally after many attempts at finding a solution, and sometimes at random moments. They sometimes come into focus as almost impossible, or perhaps wrong-headed. 

Typically, in something completely uncharacteristic for me, when I find I can’t articulate what I’m feeling or seeing, I know I’m on to something, and that’s the moment to ask “why not?” rather than “why?”

I’m also very familiar with the desire or temptation to do something new instead of doing again what I know works. Speaking from my experience that’s yielded both positive and negative results. There’s a reason some things are called “tried and true.” There’s also a reason to hold your breath, roll the dice and gamble it all on something new.

Honestly, either way is a risk. And that’s how it should be. But if you feel the need for change, go for it and don’t reverse course. 

But what do I know? I’m just a theatre guy who’s produced hits and flops along the way and comfortable taking slings and arrows along with occasional accolades. I’m not sure what feels better, being admired for a courageous leap of faith, or feeling accomplished for sticking the landing. In the end, I’m not sure it matters.

Liquid Glass 

A lot has already been said, good, bad, or indifferent about how Apple’s designers dreamed up its new Liquid Glass design approach. But that doesn’t answer the “why?” Was it a compulsion for something new? Time for a change? A diversion to distract? Or a romantic new vision spurred on by a heavy new headset?

App developer and designer Craig Hockenberry of the

, in an interesting post recently asked that question and provided what he thinks is a possible answer. It’s titled simply Liquid Glass. Why? I don’t want to spoil the post. You should go read it yourself. But his answer points to a possible future of devices “with screens that disappear seamlessly into the physical edge.” 

A cautionary note here. For several years Apple trumpeted “edge-to-edge screens” that still had bezels. Marketing mavens often outrace product dreamers to the destination.

I joked with Craig on Mastodon that he should have subtitled the piece Liquid Glass is an Edge Case. 

The joke may indeed prove to be true, but it’s a truth we’ll live with in some form or fashion for the next few years, edge case or not, regardless of the good, bad, or indifferent reactions.

Anyway, go read Craig’s piece. However Liquid Glass is received in a few weeks, I’m looking forward to discovering it myself.

I mean, why not?

You can also find more of my writings on a variety of topics on Medium at this link, including in the publications Ellemeno and Rome. I can also be found on social media under my name as above. 

Apple Brings Back Blood Oxygen Feature to the Apple Watch

Apple breathing easier with friends in high places

Apple Watch users can breathe easier again. Or at least measure their blood oxygen levels with their Apple Watches.

Lcimg 25fea456 08ef 45e8 9b0e bab9eb7f8b13.

Apple today announced that an update to Apple Watches and iPhones (16.1) to be released today will bring a redesigned Blood Oxygen feature for those with the Series 9, Series 10, and Apple Watch Ultra 2 users.

This comes after a long legal fight with Masimo, a health technology company, over a patent dispute that required Apple to deactivate the feature on Apple Watches in the U.S. that included the feature. Apple had to stop selling Apple Watches without the feature intact, but got a pause on that import ban for the Christmas season in 2023.

The patent case is still under appeal by Apple, but Apple is saying that the move comes after a U.S. Customs ruling that allows Apple to once again import Apple Watches with the Blood Oxygen feature.

I may be speculating, (I doubt it) but it sounds like a friends in high places who like gifts moment to me.

You can also find more of my writings on a variety of topics on Medium at this link, including in the publications Ellemeno and Rome. I can also be found on social media under my name as above. 

Change Is Hard

Why hasn’t AI figured this out yet?

Change may be inevitable but change is hard. Change becomes harder when those making the change, for whatever reasons, don’t remember change is hard. The only thing that doesn’t change is how easily we forget that change is hard.

Linus nylund Q5QspluNZmM unsplash.

OpenAI met with some real friction after announcing its big changes last week. Apple is going to meet some when it doles out its new operating systems with Liquid Glass next month. HBO changes its name so often it can’t even get it right in press releases. The list is as long as history. Every company faces this. Some do it well. Others not so.

As  M.G. Siegler points out in this column if you’ve been around long enough you learn to recognize the patterns. You have to be willfully blind or consumed by ego not to. In fact, the problems with instituting change are so predictable it makes one wonder why these AI engines, endlessly regurgitating whatever human wisdom they can scrape and scrounge, don’t caution against it. I’m sure somewhere in all the words and wisdom created by humans “change is hard” has been said before.

If we’re marching towards an advanced AGI with PhD level knowledge that can reason better than humans, I think the masters of the AI universe need to solve that problem before anyone can make a claim that we might someday get there.

Call me when that happens.

It’s like watching a new edition to the Alien franchise hoping one actually turns out to be more than a repeat. Or watching an American football team with a bad offensive line try to run the ball up the middle over and over again. Or thinking that once inflation retreats that prices will come down. Or thinking humans will one day be smart enough not to fall for obvious con games.

The unsolvable riddle about change involves the variables and vagaries of human nature. That’s a constant that will never change.

You can also find more of my writings on a variety of topics on Medium at this link, including in the publications Ellemeno and Rome. I can also be found on social media under my name as above. 

(Image from Linus Nylund on Unsplash)